8000 GitHub - theorizchy/rf_anomaly_detection: Machine Learning based RF Electromagnetic emanation detection
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

theorizchy/rf_anomaly_detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Overview

rf_anomaly_detection is a real-time hidden camera detector. We are using a Deep Learning model, processing captured RF activity generated by Electromagnetic Emanation produced by the hidden camera.

Setup needed

  • RaspberryPi 4B or ESP-32 Cam
  • Raspi Camera 2.1
  • Keyboard and mouse to interface with RaspberryPi Camera
  • SignalHound BB60C to capture RF activity with antenna

Follow these steps to start using rf_anomaly_detection

1. Create a python virtual environment

python -m venv venv

2. [OPTIONAL] Create an alias for virtual-environment activation

nano ~/.bashrc
alias activate='source venv/bin/activate'
source ~/.bashrc
activate

3. Install all the required dependencies

pip install -Ur requirements.txt

4. Download SPIKE software

Go to https://signalhound.com/spike/ and download the suitable file format

Architecture Operating System Header 3
32-bit Windows 7,8,10,11 Click Here
64-bit Windows 7,8,10,11 Click Here
64-bit Ubuntu 22.04/20.04/18.04 Click Here

5. Connect SignalHound (Tested with BB60C)

6. Launch the spike.exe

7. Start real-time anomaly detection

python reaaltime_anomaly_detector.py

8. Generated files

The generated output logs will be stored under "output_file/". Each file will starts with "real_time_SWEEP_"

Sample output
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 89.033%
Saved data and prediction at 2024-08-15 17:22:45.588995, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 17ms/step
1/1 [==============================] - 0s 12ms/step
1/1 [==============================] - 0s 19ms/step
1/1 [==============================] - 0s 16ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 88.322%
Saved data and prediction at 2024-08-15 17:22:45.779757, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 18ms/step
1/1 [==============================] - 0s 12ms/step
1/1 [==============================] - 0s 17ms/step
1/1 [==============================] - 0s 2ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 88.354%
Saved data and prediction at 2024-08-15 17:22:45.962733, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 17ms/step
1/1 [==============================] - 0s 11ms/step
1/1 [==============================] - 0s 21ms/step
1/1 [==============================] - 0s 6ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 86.865%
Saved data and prediction at 2024-08-15 17:22:46.149401, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 7ms/step
1/1 [==============================] - 0s 16ms/step
1/1 [==============================] - 0s 16ms/step
1/1 [==============================] - 0s 16ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 83.934%
Saved data and prediction at 2024-08-15 17:22:46.337526, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 13ms/step
1/1 [==============================] - 0s 12ms/step
1/1 [==============================] - 0s 3ms/step
1/1 [==============================] - 0s 15ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 86.757%
Saved data and prediction at 2024-08-15 17:22:46.515428, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 18ms/step
1/1 [==============================] - 0s 13ms/step
1/1 [==============================] - 0s 13ms/step
1/1 [==============================] - 0s 6ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 86.530%
Saved data and prediction at 2024-08-15 17:22:46.695669, camera: ON (confidence: 100.000%)
1/1 [==============================] - 0s 5ms/step
1/1 [==============================] - 0s 12ms/step
1/1 [==============================] - 0s 7ms/step
1/1 [==============================] - 0s 10ms/step
[KNOWN] The new camera data is most similar to the camera_1 with similarity score: 86.952%
Saved data and prediction at 2024-08-15 17:22:46.885549, camera: ON (confidence: 100.000%)
End of simulated data. Exiting...
Number of predictions=0: 0
Number of predictions=1: 99
   - [camera_1]: 99
Time taken for predicting camera operational state:
   - min: 35.418 ms
   - avg: 58.338 ms
   - max: 450.114 ms
Time taken for distinguishing camera signature:
   - min: 102.999 ms
   - avg: 122.926 ms
   - max: 250.298 ms
(venv) 

About

Machine Learning based RF Electromagnetic emanation detection

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published
0